Journal on Policy and Complex Systems
all cases were run only 10 times . The random seed value of these 10 runs was the same across cases . The improvement of computational efficiency will be one of our future issues .
The main evaluation index in our simulation results was the total number of agents , the proportion of people with immigrant traits , the frequency of mtD- NA haplogroup D4 , and the proportion of the immigrant-style pottery across all agents 400 years later .
For the total number of agents , we investigated how much the population increased among cases . For the proportion of people with immigrant traits , 80 % was considered a measure of demographic transition . For the frequency of mtD- NA haplogroup D4 , with reference to Shinoda , Kakuda , and Doi ( 2013 ), the value of 35 % is regarded as the results along the anthropological facts . For cases that met these requirements , we depicted a time series of the compositional proportion of each descendant of the agriculture holders to investigate who had performed agriculture .
Additionally , to discuss the pottery-style diffusion patterns , among cases of only vertical transmission and cases of vertical and horizontal transmission , we investigated how much the proportion of immigrant-style pottery is high . Furthermore , in cases in which the proportion of immigrant-style pottery was more than ~ 80 %, we investigate how to diffuse the immigrant-style pottery by depicting a time series of the compositional proportion of each descendant of the immigrant-style pottery holders .
3 . Results and Discussion
3.1 Difference in the Total Number of Agents Depending on the Speed of Agricultural Diffusion
The total number of agents varied depending on the speed of diffusion of agriculture . Cases in which the speed of agricultural diffusion is rapid ( e . g ., a wide [ five cells ] range of diffusion cells and an easy introduction rate [ 1 %]) had a larger total number of agents than cases in which the speed of agricultural diffusion is slow ( e . g ., a narrow [ one cell ] range of diffusion cells and a difficult introduction rate [ 0.1 %]) ( Figure 1 ). Considering these , we generated a hypothesis that the reason for population explosion in the Gusuku period is that agriculture had been diffused quickly and widely . The results coincide with our intuitions ; however , the interpretations of the results should be further investigated , whether they are a possible simulation artifact or it is a natural outcome .
78